Saturday, March 03, 2007

the machine is not yet us

"We are the web ... we are teaching the machine ... we teach it an idea"

The Machine is Us/ing Us by Michael Wesch, Assistant Professor of Cultural Anthropology, Kansas University, is currently the most popular video on YouTube

I enjoyed it too and had similar enthusiasm when I wrote about the original Kevin Kelly article on which it is based, in 2005

There are some important underlying ideas in there about the separation of style and content in XML and how this enables automatic data exchange which can be aggregated and programmed into the semantic web. The Michael Wesch video adaptation is a dynamic, engaging production that draws you in

But something is not quite right here, something which I missed earlier

Both the Wesch video and the original Kevin Kelly article go beyond saying that the internet is an intelligent assistant. They are saying that with more of the same, us humans teaching the machine, that the machine itself will turn into an autonomous thinking intelligence. This is more hinted at in the video but is explicit in the Kevin Kelly article:
It has already surpassed the 20-petahertz threshold for potential intelligence as calculated by Ray Kurzweil. For this reason some researchers pursuing artificial intelligence have switched their bets to the Net as the computer most likely to think first ... We will live inside this thing ...

The human brain has no department full of programming cells that configure the mind. Rather, brain cells program themselves simply by being used. Likewise, our questions program the Machine to answer questions. We think we are merely wasting time when we surf mindlessly or blog an item, but each time we click a link we strengthen a node somewhere in the Web OS, thereby programming the Machine by using it
- We Are the Web
They are saying that the internet is like a human brain and that when we tag we are bringing that brain to life. I don't think that's right.

I take Rodney Brooks as a guide here, that the Ray Kurzweil singularity idea is too simple. That we need more conceptual breakthroughs about human cognition and how it works before we can make predictions such as the above. Brooks challenges the idea that exponential growth in itself will automatically produce the dramatic changes envisaged by Kurzweil.

For example, AI has been working on generic object recognition for 40 years but still can't do it.

We don't have a conceptual model of how the brain works. Theoretical, conceptual breakthroughs are required. Growth itself, even though exponential, is not sufficient.
A long time ago the brain was a hydrodynamic system. Then the brain became a steam engine. When I was a kid, the brain was a telephone switching network. Then it became a digital computer. And then the brain became a massively parallel digital computer. About two or three years ago I was giving a talk and someone got up in the audience and asked a question I'd been waiting for — he said, "but isn't the brain just like the World Wide Web?"

The brain is always — has always been — modeled after our most complex technology. We weren't right when we thought it was a steam engine. I suspect we're still not right in thinking of it in purely computational terms, because my gut feeling is there's going to be another way of talking about things which will subsume computation, but which will also subsume a lot of other physical stuff that happens.
- Rodney Brooks

Connection is not everything and progress requires more than more tagging. We still need deep analytical human thought and breakthroughs to work out the future.


Artichoke said...

Thanks for this analysis Bill, I shared the video with my cluster teachers but didn't pick the reason for my disquiet - Connection is not everything and progress requires more than more tagging. buttons the button

Tony Forster said...

These discussions remind me of the plot to the movie Terminator, the defence computer network became sentient and realised that humans represented the greatest danger to its existance.

If anything we have today has the potential to become sentient, its the internet.

Bill Kerr said...

hi arti,

yes i had that feeling of disquiet too which wouldn't go away - perhaps my analysis could be further improved though, look at the list of things that Michael Wesch asks us to rethink at the end:

This touches a nerve and I think maybe he is right, we will have to think all these things

For example, my observation and participation in your blog and other blogs has brought to life the idea (which perhaps originated with vygotsky) that through language and argumentation - through comments - we literally author and re-author ourselves as humans. So when Wesch puts "identity" on the list I have to agree - the blogosphere augments vygotsky's insight manyfold

But perhaps Wesch doesn't go into the reasons for this necessary rethinking deeply enough

Nevertheless, the very clever shuffling of text around on the screen has a hypnotic effect that draws you in. Is the medium being more powerful than the message in this case?

Bill Kerr said...

hi tony,

I'd like to explore this concept of sentience and whether the internet could become sentient

Great final scene in the Terminator when the future internet turns against its creator ... I think this could only happen with the help of intelligent robots, the internet couldn't do it on its own because it doesn't have a body or perception. The internet is a dumb network or hollow sphere with all the intelligence supplied at the edges.

Sentience requires self consciousness and emotional systems. Contrast this with the semantic web, which is an Expert System. See The Semantic Web , Scientific American article by Tim Berners-Lee

I think the machine evolution of sentience will require robots that are situated and embodied in the real world.

It's a bit like the brain in a vat thought experiment. Could our brain exist in a vat after our bodies die and still think? I think not because we require our bodies and perception system to remain grounded and in touch with the world. These ideas are discussed a bit more on the Rodney Brooks page of Learning Evolves.

Artichoke said...

it is the rethinking of "authorship" that first captured my thinking when i began trying to figure out how computer mediated stuff was different from say print mediated stuff.

The thinker who has helped me with this most is Lev Manovich on notions of "collaborative authorship" and I have never strayed too far from his analysis in "The Language of New Media" 2001 since